Comparative genomic analysis demonstrates that true reinfection following SARS-CoV-2 infection is possible

Background: In recent months, multiple cases of confirmed SARS-CoV-2 reinfection have been reported. However, accurate epidemiological and virological data, including genomic analysis where possible, are required to differentiate cases of prolonged viral RNA shedding (i.e. intermittent detection) from true reinfection. The objective of this review was to systematically identify and summarise all cases of SARS-CoV-2 reinfection confirmed by comparative genomic analysis. Methods: A protocol based on Cochrane rapid review methodology was employed. Databases and pre-print servers were searched until 9/11/2020. Results: Ten studies, representing 17 patients, were identified (mean age=40; 71% male). The time interval between primary infection and reinfection ranged from 13 to 142 days (median: 60). Comparative whole genome sequencing confirmed reinfection in 14 patients (the primary and secondary infections were caused by different viruses). A further three cases had strong, but not confirmed evidence of reinfection, as only partial genomes were retrieved on primary infection. Across 12 studies that reported the number of single nucleotide polymorphisms (SNPs) comparing the first and second genomes, between 8 and 24 SNPs were discovered. With an average SARS-CoV-2 mutation acquisition rate of 1-2 per month, in all cases it is likely that the secondary infection was caused by a different SARS-CoV-2 virus, rather than prolonged shedding of viral RNA from the primary infection. In five reinfection cases, the primary and secondary infections were caused by different SARS-CoV-2 lineages/clades, strongly indicating that infections were caused by different viruses. Conclusion: Comparative genomic analyses from 14 patients confirm that SARS-CoV-2 reinfection can occur.